Computer Vision-Based Construction Error Detection Method for Glass Curtain Walls

Abstract

The present invention relates to the field of machine vision technology, and more particularly, to a method for detecting construction errors of glass curtain walls based on computer vision. The method includes: Step 1: Obtain the original image of the glass curtain wall through a camera; acquire the surface reflection characteristics of the glass curtain wall using polarization imaging technology; perform intensity reflection compensation on the original image based on the surface reflection characteristics to obtain a compensated image; Step 2: Extract feature vectors from the compensated image, and calculate the matching metric between each point in the compensated image according to the feature vectors; Step 3: According to the matching metric between each point, combined with the camera calibration parameters, perform two-dimensional reconstruction correction to obtain the corrected value of the two-dimensional point coordinates corresponding to each point; Step 4: Compare the corrected values of the two-dimensional point coordinates with the design standard template, analyze the deformation and evaluate the error. The present invention makes the error evaluation more accurate.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine vision, and particularly relates to a method for detecting construction errors of glass curtain walls based on computer vision. Background Art

[0002] As an important part of modern architecture, glass curtain walls are widely used in high-rise buildings, commercial complexes, and landmark buildings. Its main advantages include good light transmittance, aesthetics, and excellent heat insulation and sound insulation performance. However, due to the special nature of the glass curtain wall structure, the control of its construction errors has a crucial impact on the overall quality, safety, and aesthetics of the curtain wall. During the actual construction process, due to installation deviations, material processing errors, temperature stress deformation, and the influence of the structural support system, the glass curtain wall may undergo local or overall deformation. If such errors are not effectively controlled, problems such as poor sealing, reduced structural stability, and increased light pollution may occur. Therefore, how to efficiently and accurately detect the construction errors of glass curtain walls has become a key technical problem in the quality control of curtain wall projects.

[0003] Traditional manual measurement mainly relies on measuring tools such as level gauges, steel tapes, and theodolites. Construction workers measure the key nodes and reference lines of the curtain wall manually and compare them with the design standards. Although this method has certain applicability, it has the following main problems: Limited measurement accuracy: Manual measurement depends on the experience of construction workers and is easily affected by environmental factors (such as wind force, lighting conditions, etc.), resulting in relatively large measurement errors. Usually, the error is at the millimeter level, and it is difficult to meet the high-precision requirements. Low efficiency: For large-area curtain walls, manual measurement takes a long time and has limited measurement points, making it difficult to achieve a comprehensive inspection of the entire curtain wall. Difficult data processing: Manual measurement data is usually recorded in tables, making it difficult to form an intuitive error distribution map, which is not conducive to subsequent analysis and construction adjustment.

[0004] With the development of optical measurement technology, laser scanning measurement has been gradually applied to the detection of curtain wall construction errors. This method uses a laser scanner to collect point clouds on the surface of the curtain wall, and realizes error analysis by calculating the deviation between the curtain wall point cloud data and the design model. This method has high measurement accuracy (up to the millimeter level) and can obtain the overall morphological information of the curtain wall. However, there are still the following problems with this technology: High equipment cost: Laser scanners are expensive and have high usage costs, making them unsuitable for small and medium-sized projects or projects with limited construction budgets. Complex data processing: A large amount of point cloud data generated by laser scanning requires complex post-processing, including processes such as denoising, registration, and reconstruction. The computational workload is large, and high requirements are placed on the computing equipment at the construction site. Prone to glass reflection interference: Glass curtain walls have the characteristic of high reflectivity, resulting in signal loss or interference of laser scanning on the glass surface, affecting the accuracy of the data. Especially in the case of complex lighting conditions, it is difficult to ensure the stability of the measurement results. Summary of the Invention

[0005] The main purpose of the present invention is to provide a method for detecting curtain wall construction errors based on computer vision, which reduces the specular reflection interference of the glass curtain wall through polarization imaging, uses a high-precision feature matching algorithm to ensure the stability of the point positions, combines structured light measurement to calculate depth information, and adopts an optimized error weight calculation model to make the error evaluation more accurate. Compared with the prior art, the present invention not only improves the detection accuracy of curtain wall construction errors, reduces the influence of ambient light and material reflection on the measurement, but also improves the calculation efficiency and can adapt to different types of glass curtain wall materials.

[0006] To solve the above problems, the technical solution of the present invention is realized as follows:

[0007] A method for detecting curtain wall construction errors based on computer vision, the method comprising:

[0008] Step 1: Obtain the original image of the glass curtain wall through a camera; use polarization imaging technology to obtain the surface reflection characteristics of the glass curtain wall; perform intensity reflection compensation on the original image based on the surface reflection characteristics to obtain a compensated image;

[0009] Step 2: Extract feature vectors from the compensated image and calculate the matching metric between each point in the compensated image according to the feature vectors;

[0010] Step 3: According to the matching metric between each point, combined with the camera calibration parameters, perform two-dimensional reconstruction correction to obtain the corrected two-dimensional point coordinate value corresponding to each point;

[0011] Step 4: Compare the corrected two-dimensional point coordinate values with the design standard template, analyze the deformation and evaluate the error.

[0012] Furthermore, the surface reflection characteristics of the glass curtain wall are obtained through the following formula:

[0013] ;

[0014] where, represents the surface reflection characteristics of the glass curtain wall at the incident angle and the phase angle ; represents the refractive index of air; represents the refractive index of the glass curtain wall, and its value range is from 1.5 to 1.7; represents the incident angle of light; represents the phase angle; represents the angle of the main axis of polarized light.

[0015] Furthermore, through the following formula, intensity reflection compensation is performed on the original image based on the surface reflection characteristics to obtain a compensated image:

[0016] ;

[0017] where, represents the intensity of the compensated image at any point ; is the intensity of the original image at any point ; is the reflectivity of the internal interface of the glass curtain wall; is the intensity of the ambient light at any point ;

[0018] Furthermore, in step 2, the feature vector is extracted from the compensated image through the following formula:

[0019] ;

[0020] where, represents the Gaussian filter with scale ; represents gradient; represents the feature vector at any point ;

[0021] Furthermore, in step 2, the matching metric between each point in the compensated image is calculated through the following formula:

[0022] ;

[0023] where, represents the feature vector at point ; represents point The eigenvector at Indicates the intensity of the compensated image at point ; Indicates the intensity of the compensated image at point ; Indicates the L1 norm operation; Indicates the sensitivity parameter for image intensity matching; if the glass curtain wall is high-transparency flat glass, The value range is from 8 to 12; if the glass curtain wall is frosted glass, The value range is from 12 to 18; if the glass curtain wall is coated glass, The value range is from 15 to 20; if the glass curtain wall is colored glass, The value range is from 18 to 25; Indicates point and point 's matching metric; by calculating the mean of the matching metrics between all points, the average matching metric is obtained.

[0024] Furthermore, step 3: According to the matching metrics between each point, combined with the camera calibration parameters, perform two-dimensional reconstruction correction to obtain the corrected two-dimensional point coordinates corresponding to each point:

[0025] ;

[0026] Among them, is the X-axis coordinate of any point ; is the Y-axis coordinate of any point ; Indicates the depth value; is the X-axis coordinate of the principal point of the camera; Indicates the Y-axis coordinate of the principal point of the camera; Indicates the position of the camera optical center; Indicates the position of the projector; Indicates the camera focal length.

[0027] Furthermore, the depth value is calculated by the following formula:

[0028] ;

[0029] ;

[0030] ;

[0031] Among them, Indicates the angle between the camera line of sight and the optical axis; Indicates the angle between the projector line of sight and the projection axis; Represents the baseline distance between the camera and the projector; Represents the structured light wavelength.

[0032] Furthermore, in step 4, according to the following formula, based on the correction value of the two-dimensional point coordinates, compare with the designed standard template, analyze the deformation and evaluate the error:

[0033] ;

[0034] ;

[0035] Among them, Represents the standard value of the designed standard template at the point with coordinates ; Is the difference value; Is the error value; Is the weight value of the point with coordinates ;

[0036] Furthermore, the weight value Is represented by the following formula:

[0037] ;

[0038] Among them, Is the attenuation parameter of the spatial weight, which is a set value; Is the sensitivity parameter of the feature intensity weight, which is a set value; ; Is the central value in the designed standard template.

[0039] A method for analyzing the force on the joints of a spherical reticulated shell structure based on refined simulation optimization of the present invention has the following beneficial effects: The high reflectivity of glass curtain walls often causes strong light interference during the imaging process, affecting the accuracy of feature extraction. The present invention uses polarization imaging technology to obtain the reflection characteristics of the curtain wall surface and uses a mathematical model to compensate the reflection intensity of the image, thereby significantly reducing the influence of specular reflection, improving the image quality, and enabling stable extraction of feature points. Traditional computer vision methods have low matching accuracy in the glass curtain wall environment, while the present invention uses a feature extraction method based on gradient information and Gaussian filtering and combines matching metric calculation to improve the stability and accuracy of point matching, ensuring that subsequent error calculations are based on high-quality data. Calculate the optical reflection characteristics of the curtain wall surface through a mathematical model and compensate the reflection intensity of the image based on this model to ensure that the system can still obtain high-quality feature images even under different lighting conditions. Traditional binocular stereo vision methods are prone to failure in the glass curtain wall environment, while the present invention combines structured light projection, calculates the depth information of points using matching metrics, and optimizes the calculation through the geometric relationship between the camera and the projector to ensure that accurate depth data can be obtained even in a highly reflective environment. In error calculation, a spatial attenuation weight based on Gaussian distribution is introduced to ensure that the error weight in key areas (such as the curtain wall edge and splicing) is higher, while the weight of flat areas with less influence on errors is lower, thereby optimizing the overall error evaluation result. Description of the Drawings

[0040] Figure 1 It is a schematic flowchart of a method for analyzing the force on the joints of a spherical reticulated shell structure based on refined simulation optimization provided by an embodiment of the present invention. Detailed Embodiments

[0041] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0042] Example 1, refer to Figure 1 : A method for detecting construction errors of glass curtain walls based on computer vision, the method comprising:

[0043] Step 1: Obtain the original image of the glass curtain wall through a camera; obtain the surface reflection characteristics of the glass curtain wall using polarization imaging technology; perform intensity reflection compensation on the original image based on the surface reflection characteristics to obtain a compensated image;

[0044] The main component of the glass curtain wall is silicate material, whose surface is smooth and has a high reflectivity. When light is incident on the glass surface, partial reflection and partial transmission will occur. This reflection can be divided into two forms: one is specular reflection, that is, light is reflected regularly according to the law of reflection; the other is diffuse reflection, that is, due to the microscopic inhomogeneity on the glass surface, part of the light is reflected in a non-regular direction. The effects of these two reflections on the computer vision system are very different. Diffuse reflection is the effective information that computer vision technology can utilize because it can truly reflect the physical form of the glass surface, while specular reflection is the interference information that needs to be compensated and removed. Since the reflection characteristics of the glass curtain wall change with the incident angle and the refractive index of the material, when performing error detection, it is first necessary to accurately obtain the reflection characteristics of the glass curtain wall and perform reflection compensation on the image based on this, so as to obtain more reliable visual information.

[0045] In this method, polarization imaging technology is introduced to effectively separate the specular reflection component and the diffuse reflection component of the glass curtain wall. When light is reflected on the glass surface, the direction of the electric field vibration of part of the light wave will change, making the reflected light show certain polarization characteristics. Using this physical phenomenon, images under different polarization directions can be obtained through a polarization camera or an additional polarization filter, and the reflectivity of the glass curtain wall under specific incident angles and phase angle conditions can be calculated through a specific mathematical model. The polarization filter usually adopts a variable angle adjustment mode to obtain the reflection conditions of the glass curtain wall at multiple polarization angles, so as to establish a complete reflection characteristic model. This model can describe the reflection behavior of the glass under different incident conditions and provide the necessary physical parameters for subsequent image compensation.

[0046] After the polarization imaging obtains the surface reflection characteristics of the glass curtain wall, the computer vision system needs to perform intensity reflection compensation on the original image based on these characteristics to remove the interference of specular reflection, so that the main information retained in the image is mainly the real information of the curtain wall surface. The core idea of compensation is to establish a mathematical model to separate the specular reflection component and restore the diffuse reflection information of the curtain wall surface in a reasonable way. Since the intensity of specular reflection is closely related to the refractive index of the glass, the incident angle, and the wavelength of light, the computer vision system needs to combine these parameters and use the physical compensation formula to correct each pixel point in the image. Specifically, the true intensity of each pixel can be restored to its proper illumination information by subtracting the part affected by specular reflection and normalizing it in combination with the reflectivity of the internal interface of the glass. In this way, the computer vision system can effectively suppress specular reflection, improve the contrast and clarity of the image, and provide more reliable basic data for subsequent feature extraction and error detection. The polarization imaging technology can separate the effective information of the curtain wall in a complex lighting environment, enabling the computer vision system to accurately perceive the true shape of the curtain wall even in a strong reflection background. In addition, traditional image de-reflection methods usually rely on multi-angle imaging or high-dynamic-range imaging, which not only increase the complexity of the system but also are difficult to apply in the glass curtain wall detection scenario. The present invention adopts a method of combining polarization imaging with physical modeling, which only requires single-view imaging to achieve efficient reflection compensation, enabling the error detection system to work stably in an ordinary construction site environment. In addition, this method performs image compensation by utilizing the physical and optical characteristics of the glass curtain wall, avoiding the uncertainty that may be introduced by the data-driven model and improving the versatility and robustness of the system.

[0047] Step 2: Extract the feature vectors from the compensated image and calculate the matching metrics between each point in the compensated image according to the feature vectors;

[0048] First, after reflection compensation, there may still be local illumination changes and residual reflection effects in the glass curtain wall. To ensure the stability of feature extraction, the system needs to enhance edge features based on the gradient information of the image. Gradient calculation is one of the basic means of feature extraction in computer vision, which can effectively capture the edge information of the image, that is, the area where the pixel intensity changes drastically. In the scenario of a glass curtain wall, the edge information often corresponds to the structural features of the curtain wall, such as the joints of glass panels, the fixed frames of the curtain wall, and local minor deformations. Therefore, by calculating the gradient magnitude and direction of the image, the basic features of the glass curtain wall surface can be obtained, enhancing its recognizability in the matching calculation. After obtaining the basic gradient features, in order to further enhance the robustness of the system, Gaussian filtering needs to be introduced for smoothing. The surface of the glass curtain wall may generate minute noises due to illumination, pollutants, or processing errors. If these noises directly participate in feature extraction, they may affect the matching accuracy. The role of the Gaussian filter is to perform scale-space processing on the image, enabling the system to stably extract features at different scales and reducing the influence of local noises. Especially in the glass curtain wall scenario, considering the different reflection characteristics of different materials (such as transparent glass, frosted glass, coated glass), Gaussian filtering can play a certain role in normalization, enabling the surface features of curtain walls made of different materials to be better uniformly processed. In addition, since the edge feature intensity of the glass curtain wall is greatly affected by the reflectivity, when calculating the gradient features, it is necessary to normalize the feature values in combination with the reflection characteristic parameters to avoid the non-uniformity of feature extraction caused by different reflectivities of different materials.

[0049] After feature extraction is completed, the system needs to calculate the matching metrics between various points in the image to establish feature relationships at the pixel level. The matching metric is an important method in computer vision for measuring the similarity between two pixel points or regions. In the glass curtain wall error detection task, the calculation of the matching metric directly determines the accuracy of subsequent reconstruction and correction. Therefore, its calculation method needs to be optimized in combination with the characteristics of the glass curtain wall material. The calculation of the matching metric is usually based on two core principles. One is the similarity of feature vectors, and the other is the similarity of pixel intensities. The similarity of feature vectors can be calculated through the cosine similarity metric to ensure that the extracted feature vectors can maintain a certain degree of alignment in both direction and amplitude. The similarity of pixel intensities, on the other hand, needs to be measured based on the Gaussian model to measure the impact of the brightness difference between two points on the matching metric. To adapt to different types of glass curtain wall materials, the method of the present invention introduces different sensitivity parameters in the calculation of the matching metric. The transparency, surface roughness, and coating conditions of the glass curtain wall will all affect the calculation of the matching metric. Therefore, during the calculation process, corresponding matching parameters need to be set according to different glass materials. For example, in the case of a highly transparent glass curtain wall, the matching metric needs to pay more attention to local texture information, while in the case of a frosted glass or coated glass curtain wall, more attention needs to be paid to the overall structural information. Therefore, by adaptively adjusting the matching parameters, this method can ensure high-precision feature matching on different types of glass curtain walls.

[0050] Step 3: According to the matching metrics between each point, combined with the camera calibration parameters, perform two-dimensional reconstruction and correction to obtain the corrected two-dimensional point coordinates corresponding to each point;

[0051] During the imaging process in computer vision, any real-world object is projected onto a two-dimensional image plane through a camera, and this imaging process is affected by the internal and external parameters of the camera. The internal parameters of the camera include focal length, principal point coordinates, distortion coefficients, etc., while the external parameters involve the rotation and displacement of the camera. Since the construction error detection of glass curtain walls relies on precise coordinate calculations, in order to ensure the detection accuracy, the system needs to compensate for and correct these effects. Especially in the scenario of glass curtain walls, there may be non-linear distortions during the imaging process, such as lens distortion, projection deviation, and optical refraction. If the coordinate correction is not performed, it may lead to positioning deviations of some key points during the error assessment process, thus affecting the reliability of the error calculation. Therefore, the system first needs to perform distortion correction on the original image using camera calibration data to remove the geometric distortion caused by lens deformation, so that the points in the image can more accurately reflect the true projection position of the curtain wall. After completing the distortion correction, the system needs to combine the matching metric to optimize the coordinate correction of the points. The matching metric is a mathematical index used to measure the similarity between various points in the image, and it can help the system determine which points are stable and reliable, and which points may drift due to factors such as lighting or reflection. In the scenario of glass curtain wall construction error detection, due to the influence of the material of the curtain wall surface, installation accuracy, and ambient lighting, the images obtained by the camera may have a certain degree of distortion and deviation. Therefore, during coordinate correction, it is necessary to focus on the points with low matching degrees and use the neighboring points with high matching degrees around them for interpolation correction to make the point coordinates of the entire image more consistent, thereby improving the accuracy of the error calculation. The calculation of the matching metric usually combines the similarity of feature vectors and the similarity of pixel intensities. Through these metric methods, it can effectively judge which points in the image can still maintain high stability after transformation, and which points need further adjustment. In addition to using the matching metric to optimize the coordinate correction, this method also combines the projection relationship of the camera to perform precise geometric transformation on the points. Since the structure of the glass curtain wall is usually relatively regular, and the construction errors generally manifest as local deformations or overall displacements, the projection transformation model can be used to adjust the coordinates to ensure that the distribution of the points in the image is more in line with the actual geometric shape of the curtain wall. In computer vision, projection transformation is a mathematical method that maps a point from the original coordinate system to another coordinate system, and it can adjust the point coordinates in the image based on the imaging model and calibration parameters of the camera, making its position closer to the true-world curtain wall structure. In the implementation of this method, the core of the projection transformation is to use the point pair information calculated by the matching metric, and through the optimal transformation calculation, make all points as aligned as possible in the transformed coordinate system, thereby improving the accuracy of the error detection.

[0052] Step 4: Compare with the design standard template according to the two-dimensional point coordinate correction value, analyze the deformation and evaluate the error.

[0053] The design standard template of a glass curtain wall is a predefined ideal geometric model, which is usually generated based on precise data from the architectural design plan and contains key information such as the standard dimensions, installation angles, joint spacings, etc. of each unit of the curtain wall. During the construction process, due to the influence of external environments, material errors, installation techniques, and other factors, the actually installed curtain wall often deviates from the theoretical design to a certain extent. Therefore, an error detection system is needed to determine whether the construction accuracy of the curtain wall meets the design requirements. In the method of the present invention, the basic principle of error calculation is to calculate the difference between the corrected two-dimensional point coordinates and the corresponding coordinates of the standard template, and use a mathematical model to quantify and visualize the error, enabling construction personnel to intuitively understand the deviation of the curtain wall and take necessary adjustment measures. During the process of calculating the error, the system first needs to establish an error measurement model, which is used to measure the deviation degree between the actual coordinates and the design standard coordinates of each point. The errors of a glass curtain wall can be divided into multiple dimensions, such as translational errors in the horizontal direction, offset errors in the vertical direction, and local deformation errors, etc. To improve the accuracy of error calculation, this method introduces an error calculation model based on weight distribution, which can assign different calculation weights according to the error contribution degrees of different regions to ensure that the measurement results are more in line with the actual situation. For example, in the edge regions of the curtain wall, due to the usually large stress concentration effect during the installation process, the errors may be relatively large. Therefore, when calculating the errors, higher weights need to be assigned to these regions to ensure that the system can more accurately reflect the overall error distribution of the curtain wall. For the central regions of the curtain wall, due to the strong rigid support of the materials, the errors are usually small. Therefore, the system can assign lower weights to the error calculations of these regions to reduce the calculation complexity and improve the stability of the analysis.

[0054] After the error calculation is completed, the system also needs to further analyze and visualize the error results so that construction workers can intuitively understand the error distribution and make construction adjustments accordingly. This method uses color coding to visually annotate the errors, that is, according to the magnitude and direction of the errors, different colors are used to distinguish them in the detection results, enabling construction workers to quickly judge which areas have larger errors and which areas have smaller errors. For example, for areas with larger errors, red can be used for annotation, while for areas with smaller errors, green can be used for annotation. In this way, construction workers can quickly identify the key error areas on the curtain wall surface and adjust the construction process according to the detection results to improve the overall installation accuracy of the curtain wall. In addition, during the error analysis process, the system can also evaluate the overall error trend through statistical methods to judge the distribution law of construction errors. For example, if the errors show a certain linear change trend across the entire curtain wall surface, it may indicate that there is a systematic deviation in the overall installation of the curtain wall, which may be caused by errors in the measurement reference points or offsets in the installation reference lines. If the errors show a locally irregular distribution, it may indicate that there are material deformations or uneven forces on the supporting structures in local areas of the curtain wall. Therefore, through the statistical analysis of the error distribution, the system can provide more in-depth error evaluation results for construction management personnel, enabling them to take targeted construction optimization measures to ensure that the installation quality of the curtain wall meets the design standards.

[0055] Example 2: The surface reflection characteristics of the glass curtain wall are obtained through the following formula:

[0056] ;

[0057] where represents the surface reflection characteristics of the glass curtain wall at the incident angle and the phase angle ; represents the refractive index of air; represents the refractive index of the glass curtain wall, and its value range is from 1.5 to 1.7; represents the incident angle of light; represents the phase angle; represents the angle of the main axis of polarized light.

[0058] Specifically, in the formula, represents the surface of the glass curtain wall at the given incident angle and the phase angle The reflectivity under certain conditions. It is used to describe the proportion of the intensity of the reflected light to the intensity of the incident light when light is incident on the surface of a glass curtain wall. Since glass is a typical transparent material, the reflection behavior on its surface depends not only on the incident angle of light but also on the refractive index difference between glass and air. Therefore, the formula includes the refractive indices of the materials and calculation terms, representing the refractive indices of air and glass respectively. The refractive index of air is usually taken as , while the refractive index of the glass curtain wall ranges from to . Different refractive indices affect the reflection intensity of light on the glass surface, resulting in different reflection characteristics for different types of glass curtain walls. For example, materials such as ordinary float glass, low-emissivity coated glass, or dimming glass all exhibit different reflection behaviors. The numerator part of this formula is mainly used to describe the reflection characteristics of the glass curtain wall surface under the action of different phase angles , where reflects the influence of the phase angle of light on the reflectivity. During the polarization imaging measurement of the glass curtain wall, the phase angle directly affects the intensity and polarization direction of the reflected light. Therefore, this term plays a role in characterizing the reflection characteristics of polarized light in the formula. When takes different values, the polarization direction of light changes, causing the reflectivity of the glass surface to change accordingly. In addition, the denominator part of this formula is used for normalization, ensuring that the calculation results always remain within a reasonable physical range and avoiding numerical overflow or situations that do not conform to optical laws. In addition, the last term of this formula further considers the influence of the principal axis angle of light polarization on the reflectivity. Since the glass curtain wall is usually installed at different angles in a building and the direction of the incident light changes with time, it is necessary to introduce the principal axis angle of polarized light for compensation to ensure that the calculated reflectivity can accurately match the actual optical measurement situation. When , the value of this term is the largest, indicating that the polarization direction at this time exactly matches the reflection behavior on the glass surface, thus making the reflectivity calculation more accurate. When deviates from , the reflectivity gradually decreases with the change of the angle, reflecting the optical reflection change law of the glass curtain wall under different incident conditions.

[0059] Example 3: Based on the surface reflection characteristics, the original image is intensity reflection compensated through the following formula to obtain a compensated image:

[0060] ;

[0061] Among them, represents the intensity of the compensated image at any point ; is the intensity of the original image at any point ; is the reflectivity of the internal interface of the glass curtain wall; is the intensity of the ambient light at any point .

[0062] Specifically, in this formula, represents the intensity of the compensated image, that is, after the reflection compensation process, the true illumination information at each pixel point . Due to the influence of specular reflection on the original image , the brightness in some areas increases abnormally, making it difficult to distinguish the characteristic details on the curtain wall surface. Therefore, it is necessary to correct it through computer vision methods. The basic principle of correction is to use the surface reflectivity of the glass curtain wall to calculate the contribution of the reflected light to the original image and subtract it from the image, so as to restore the true illumination distribution on the curtain wall surface. Here, is calculated by the aforementioned optical model, which describes the reflection intensity of light at different incident angles and phase angles , and determines the specular reflection contribution of the glass curtain wall to the ambient light. In this calculation process, the ambient light is an important variable, which represents the illumination intensity in the environment where the glass curtain wall is located. Since the glass curtain wall is usually in a complex outdoor environment, the ambient light source may include sunlight, skylight, reflected light from surrounding buildings, etc., so its influence is complex and difficult to predict. This method obtains the influence information of the ambient light through polarization imaging technology and estimates it using a physical model, making the compensation calculation more accurate. In the formula, this term represents the contribution of the ambient light caused by specular reflection at a certain pixel point of the glass curtain wall, which is the interference component that needs to be removed from the original image. Therefore, the numerator part of the formula calculates the remaining image information after subtracting the reflected light. However, the reflection of light on the glass curtain wall not only occurs on the surface layer, but may also involve multiple reflections at the internal interface of the glass. This is because glass is a semi-transparent medium, and part of the incident light will penetrate the surface and enter the glass interior, and be reflected at the internal interface, thus further affecting the brightness of the final image. Therefore, the denominator part of the formula is used to normalize the overall illumination distribution to compensate for the brightness deviation caused by internal reflection in the glass. At the same time, in order to further reduce the influence of reflection at the internal interface of the glass, this formula also introduces the reflectivity of the internal interface of the glass Perform corrections, where this item represents the correction of the influence of internal reflection in the glass on the overall brightness, so that the final compensated image can be closer to the true surface information of the glass curtain wall. This method not only removes reflections based on the optical properties of the glass curtain wall, but also combines the actual environmental light influence, making the compensation effect more accurate. Compared with traditional image processing methods, such as histogram equalization or adaptive filtering, which are only based on the statistical features of the image, the physical modeling method of this method can more accurately separate the reflected light component and effectively restore the true visual information of the glass curtain wall. In addition, this method can also adapt to different types of glass curtain wall materials, such as transparent glass, low-emissivity glass, coated glass, etc., because its compensation calculation is based on the physical parameters of refractive index and reflectivity, and these parameters can be adjusted according to specific materials, so as to ensure that the compensation model can be applied to the detection needs of different curtain wall structures. In practical applications, this reflection compensation model can significantly improve the accuracy of error detection, enabling the computer vision system to still accurately identify the structural features of the glass curtain wall in complex lighting environments. For example, under strong sunlight, the uncompensated image may have large brightness non-uniformity due to excessive specular reflection, resulting in a decrease in the accuracy of edge detection and feature matching. However, for the image processed by this compensation formula, its brightness distribution will be more balanced, enabling the system to more reliably extract the effective information on the curtain wall surface, thereby improving the accuracy of error detection and construction quality assessment.

[0063] Example 4: In step 2, extract the feature vector from the compensated image through the following formula:

[0064] ;

[0065] where represents the Gaussian filter with scale ; represents gradient; represents the feature vector at any point .

[0066] Specifically, in this formula, represents the feature vector calculated at any pixel point , and this feature vector is used to describe the local structural information of this pixel point on the glass curtain wall surface. In computer vision tasks, feature vectors are usually used to match and align corresponding points in different images, so their stability is crucial for subsequent matching calculations. In order to extract stable feature information, this method first calculates the gradient of the compensated image , that is . The role of gradient calculation is to highlight the edge and texture information in the image, enhance the structural features of the curtain wall surface, enabling the system to more accurately identify the boundary and deformation area of the curtain wall. Since the features of the glass curtain wall are usually sparse, directly using the gray value as the matching basis has a poor effect, while the gradient information can better describe the local changes in the image, thus improving the effect of feature extraction. However, gradient calculation itself is easily affected by noise. Especially in the glass curtain wall detection scenario, due to possible local illumination changes, pollutants, or minute texture interferences caused by construction errors on the curtain wall surface, the gradient calculation results may show over-response or local anomalies. Therefore, this method introduces a Gaussian filter to smooth the gradient calculation results. Gaussian filtering is a commonly used image processing method. It can reduce the interference of random noise while retaining the main features of the image, making the feature extraction process more robust. In this formula, the scale parameter of the Gaussian filter needs to be adjusted according to the specific application scenario. Usually, a smaller value is used to extract detailed features, while a larger value is used to extract global structural information. In the glass curtain wall error detection task, the system can adaptively adjust value according to the different materials of the curtain wall surface to optimize the effect of feature extraction.

[0067] In addition, since the reflectivity of the glass curtain wall will affect the gradient calculation results, this method also introduces a normalization term , which is used to compensate for the non-uniformity of gradient values caused by reflection. The reflectivity of glass curtain walls varies greatly at different angles. Especially in the case of large-angle incident light, the reflectivity of some areas may be close to 1, resulting in a reduced contrast of the image in these areas, thereby affecting the accuracy of gradient calculation. Therefore, the role of this normalization term is to amplify the gradient response in low-reflection areas, ensuring that even under high-reflection conditions, the system can still extract stable feature vectors. In this way, the system can adapt to different types of glass curtain wall materials, including transparent glass, coated glass, and low-emissivity glass, etc., and maintain the stability of feature extraction under different lighting environments. This feature extraction method combines the dual advantages of physical modeling and image processing, enabling the feature extraction process of glass curtain walls to not only reduce reflection interference but also enhance effective information, ensuring the stability of feature vectors in different environments. Compared with traditional feature extraction methods, such as SIFT (Scale-Invariant Feature Transform) or SURF (Speeded-Up Robust Features), this method is specifically optimized for the special optical characteristics of glass curtain walls, avoiding the influence of high-smoothness surfaces on the stability of feature point matching. At the same time, since this method is based on image gradient calculation and avoids the complex multi-scale feature extraction process, the computational amount is lower, which can meet the real-time requirements in practical engineering applications.

[0068] Example 5: In step 2, according to the following formula, the matching metric between each point in the compensation image is calculated based on the feature vectors:

[0069] where, represents the feature vector at point ; represents the feature vector at point ; represents the intensity of the compensation image at point ; represents the intensity of the compensation image at point ; represents the L1 norm operation; represents the sensitivity parameter of image intensity matching; if the glass curtain wall is high-transparency flat glass, ranges from 8 to 12; if the glass curtain wall is frosted glass, ranges from 12 to 18; if the glass curtain wall is coated glass, ranges from 15 to 20; if the glass curtain wall is colored glass, ranges from 18 to 25; represents the matching metric between point and point ; by calculating the mean value of the matching metrics between all points, the average matching metric is obtained.

[0070] Specifically, the first part of the formula is the cosine similarity calculation based on feature vectors: , and the purpose of this part of the calculation is to measure the similarity degree of the feature vectors of two points in terms of direction. Since the surface of the glass curtain wall is usually smooth and its local area may lack obvious texture information, the direct matching method based on pixel intensity may not accurately distinguish the similarity of different points. Instead, using the feature vector method can more stably extract the local structure information of the glass curtain wall surface. In this formula, and respectively represent the feature vectors at points and . And the denominator part is normalized so that the calculation result only depends on the direction of the feature vector and is not affected by the amplitude of the feature vector. This normalization method is called cosine similarity, which is widely used in feature matching and image retrieval tasks in the field of computer vision. Specifically, the value range of cosine similarity is between . When the directions of two feature vectors are completely the same, the value is 1, indicating that the local structures of the two points are highly similar; when the directions of the two vectors are completely opposite, the value is -1, indicating that they are completely unmatched; when the two vectors are orthogonal, the value is 0, indicating that there is no correlation between them. In the error detection task of the glass curtain wall, the feature vectors are usually calculated based on gradient information and Gaussian filtering. Therefore, this matching method can maintain good stability under different lighting conditions and reduce the mis-matching situation in a highly reflective environment.

[0071] However, relying solely on the similarity of feature vectors is not sufficient to ensure the accuracy of the matching calculation because the surface reflection characteristics of the glass curtain wall may cause the gradient information in some areas to be distorted, thus affecting the stability of the feature vectors. Therefore, to further enhance the robustness of the matching, this method introduces a weighted correction factor based on the image intensity difference in the matching calculation, that is: ; this term is used to measure the similarity degree of the light intensity of two points on the compensated image and adjusts the matching confidence through an exponential decay method. Intuitively, this weighted factor can be understood as: if the light intensity difference between two points on the compensated image is small, their matching metric value should be high, and if the light intensity difference between the two points is large, the matching metric value should be low. The form of the exponential function ensures that the matching degree decreases non-linearly as the light intensity difference increases, enabling the system to more effectively ignore the light intensity changes caused by local lighting variations or different glass curtain wall materials. For example, in some areas of the glass curtain wall, the local brightness may increase due to the influence of external ambient light, and this exponential function can automatically reduce the matching confidence in these areas, thereby reducing the matching error. The parameter It plays a role in sensitivity adjustment in this calculation, which determines the degree of response of the system to changes in light intensity. Under different materials of glass curtain walls, the optimal values of this parameter are different. For example, for highly transparent glass, the light mainly propagates by transmission, and the surface reflectivity is relatively low. Therefore, in the matching calculation, the system needs to be more sensitive to smaller differences in light intensity. Thus should take relatively small values (8 to 12). For frosted glass, due to its strong scattering characteristics on the surface, the local changes in light intensity are large. Therefore, the matching calculation requires a lower sensitivity to avoid excessive matching errors caused by light changes. Thus should be in the range of 12 to 18. Similarly, for coated glass, due to its relatively complex surface structure and large variation in reflection characteristics with the incident angle, should take larger values (15 to 20) to ensure the stability of the matching calculation. For colored glass, since its surface color may affect the perception of light intensity, the matching calculation requires greater robustness. At this time should be in the range of 18 to 25. By adaptively adjusting , this method can automatically optimize the matching calculation for glass curtain walls of different materials, thereby ensuring the stability and accuracy of the matching results. In addition, to evaluate the global characteristics of the entire matching calculation, this method also introduces the calculation of the mean value of the matching metric, that is, by calculating the average matching metric between all points, the matching trend of the entire curtain wall surface can be obtained. The statistical analysis of the mean value of the matching metric can be used to detect the overall construction error of the glass curtain wall. For example, when the mean value of the matching metric is relatively high, it indicates that the overall shape of the glass curtain wall is relatively consistent with the design standard. When the mean value of the matching metric is relatively low, it may mean that there are relatively large errors in some areas of the curtain wall. In addition, the spatial distribution of the mean value of the matching metric can also be used to identify local deformation conditions on the curtain wall surface. For example, if the matching metric in a certain area is significantly lower than the global mean value, it indicates that there are relatively large deviations in the structural characteristics of this area from the standard template, and local correction may be required.

[0072] Example 6: Step 3: According to the matching metrics between each point, combined with the camera calibration parameters, perform two-dimensional reconstruction correction to obtain the corrected two-dimensional point coordinates corresponding to each point:

[0073] ; where is the X-axis coordinate of any point ; is the Y-axis coordinate of any point ; represents the depth value; is the X-axis coordinate of the principal point of the camera; represents the Y-axis coordinate of the principal point of the camera; Indicates the position of the camera optical center; Indicates the position of the projector; Indicates the camera focal length; Is the transpose symbol.

[0074] Specifically, the core variables of the formula are , which represents the corrected two-dimensional point coordinates, that is, the point positions after matching metric and projective geometry correction. Since the installation errors of glass curtain walls usually affect the overall shape of the curtain wall on a relatively large scale, the error detection system needs to calculate the true position of each point as accurately as possible, so as to provide accurate coordinate information for subsequent error calculation. When calculating , the imaging characteristics of the camera need to be considered, where represents the position of the camera optical center, and is the camera focal length. The focal length affects the field of view angle of the camera and determines the perspective projection relationship in the image. In the error detection task of glass curtain walls, short focal length cameras are usually used to ensure that a large area of the curtain wall can be captured while taking into account the resolution. However, short focal length cameras are prone to more obvious perspective distortion. Therefore, when performing coordinate correction, it is necessary to use projection transformation methods to compensate for it. To achieve this compensation, the following projection scaling factor is introduced into the formula: ; The calculation of this part describes the scaling transformation relationship of point in the camera coordinate system. Among them, represents the depth value, represents the relative distance between the camera and the projector, and the denominator part represents the perspective scaling factor of point on the camera imaging plane. The role of this scaling factor is to adjust the position of the point in the image according to the projection depth of the point, so that the coordinate correction can fully consider the spatial distribution of the curtain wall point positions. In the error detection task of glass curtain walls, different points may have different projection scaling ratios due to different depths. Therefore, it is necessary to compensate through this scaling factor to ensure the accuracy of point position calculation.

[0075] Next, the last term of the formula: ; This part is used to describe the direction vector of point in the camera coordinate system. Since the error detection of glass curtain walls relies on accurate point matching, and the perspective projection relationship of the camera may cause certain distortion of the points in the image, it is necessary to calculate the true projection direction of the point coordinates through this direction vector, so that the corrected coordinates can more accurately reflect the actual position of the curtain wall. The calculation method of this direction vector is based on the optical projection model of the camera, where and respectively represent point Offset relative to the camera principal point and represents the focal length of the camera, ensuring the normalized calculation of the direction vector and making the coordinate correction process unaffected by projection distortion. The application of this formula in the task of detecting construction errors of glass curtain walls enables the system to more accurately map the actual coordinates of curtain wall points, thereby improving the accuracy of error calculation. Compared with traditional plane projection-based methods, this method combines the depth information obtained from matching calculations and uses projection transformation for coordinate correction, enabling the system to adapt to different types of glass curtain wall materials and maintain high calculation stability under complex lighting conditions. In addition, by introducing camera calibration parameters, this method enables error detection to adapt to different camera devices, thereby improving the applicability of the system. In practical applications, this two-dimensional reconstruction and correction method can significantly improve the accuracy of curtain wall error detection. For example, in ordinary construction surveys, due to possible deviations in the camera shooting angle, directly calculating errors based on the original image may lead to large coordinate offsets. After coordinate correction using this method, the system can accurately correct the points to the true coordinate system of the camera, thereby reducing the influence of perspective distortion on error detection. In addition, this method can also adapt to different types of glass curtain wall materials, such as transparent glass, low-emissivity glass, coated glass, etc., and can maintain high calculation stability under complex lighting conditions.

[0076] Example 7: Depth value is calculated by the following formula:

[0077] ;

[0078] ;

[0079] ;

[0080] where represents the angle between the camera line of sight and the optical axis; represents the angle between the projector line of sight and the projection axis; represents the baseline distance between the camera and the projector; represents the wavelength of the structured light.

[0081] Specifically, based on the principle of triangulation, this method establishes a projection geometric model with the camera optical center as the reference. Using the fixed physical distance between the camera and the projector as the baseline, a direct correspondence is established between the fringe information generated by the structured light projection and the depth of the actual object surface, thus solving the problem that traditional stereo vision methods are easily interfered with and difficult to obtain depth information on high-reflection surfaces such as glass curtain walls. In the formula, by coordinating the geometric positions of the camera and the projector, the depth is calculated The formula form is The physical meaning of this expression is to convert the actual physical baseline Combined with the sinusoidal function relationship between two angles, it constitutes a typical triangulation model.

[0082] Specifically, the projector projects a light pattern of a specific wavelength onto the glass curtain wall through structured light. Due to its special optical properties, the glass curtain wall will cause displacement and deformation of the projected pattern in different areas. After the preliminary matching measurement and feature extraction, this deformation information can reflect the offset of the pattern in the image. Using these offsets, the corresponding line of sight angle can be calculated, thereby deriving the depth of each point relative to the camera. Here, the angle is calculated by the camera imaging model, which reflects the The deviation from the camera's optical axis, which is calculated by the distance between the point and the camera's principal point and the focal length The size of this angle directly determines the scaling of the point under perspective projection. It comes from structured light technology, which measures the average matching Structured light wavelength And the baseline distance The function relationship is inversely solved to obtain the angle between the projector's line of sight and the projection axis. In fact, the advantage of the structured light system in glass curtain wall construction error detection is that it uses a light source with a known wavelength and a fixed geometric structure to measure the depth information of the object surface through the distortion of the fringe pattern, which is particularly important for glass materials with high reflectivity and low texture.

[0083] The entire depth calculation process not only requires accurate measurement of the physical distance between the camera and the projector, but also relies on accurate calibration of the camera's internal parameters, such as focal length. and the principal point coordinates The accurate calibration of these parameters is the premise of building a correct projection model. Only on this basis can the pixel position in the two-dimensional image be mapped back to the actual object surface by using the perspective projection principle, thereby achieving accurate depth reconstruction. Through the sine function involved in the formula, it can be seen that when the angle between the camera line of sight and the optical axis is When the depth changes The angle between the projector's line of sight and the projection axis is also adjusted accordingly, which reflects the size difference between near and far objects and the effect of perspective offset in perspective projection. It plays a role in adjusting the sensitivity of depth calculation. The numerator part of its sine function combines the structured light wavelength and the matching metric, enabling the depth information to remain stable under different lighting conditions and material reflection characteristics. In practical applications, due to its material properties, glass curtain walls often exhibit complex reflection phenomena and optical refraction problems, which can easily interfere with traditional depth calculation methods and result in large errors. However, by introducing structured light technology and combining it with the precise calibration of the camera and projector, the present invention can not only overcome the challenges brought by high reflectivity but also maintain high measurement accuracy in complex environments. Using this triangulation method, the depth information of each pixel point can be accurately obtained and combined with subsequent two-dimensional coordinate correction steps to achieve high-precision measurement of the entire error detection system. It should be noted that during the actual measurement process, the installation angles and relative positions of the camera and projector may deviate due to changes in construction site conditions. Therefore, the system usually needs to have self-calibration capabilities to dynamically adjust these parameters based on real-time collected data, which is also a major innovation point of the present invention. In addition, using the matching metric as the basis for calculation enables the depth calculation to adaptively reflect the subtle differences in the local structure of the glass curtain wall surface. When local deformation occurs on the curtain wall surface due to construction errors or material defects, the matching metric will show corresponding abnormal changes, which are directly transmitted to the depth calculation, enabling the system to detect depth deviations in local areas and further perform error analysis. It is precisely through this multi-level and multi-parameter combined depth calculation method that the present invention achieves high-precision evaluation of deformation and installation errors in glass curtain wall construction error detection, ensuring the scientificity and reliability of the detection results.

[0084] Example 8: In step 4, according to the following formula, compare with the design standard template based on the two-dimensional point coordinate correction value, analyze the deformation, and evaluate the error:

[0085] ;

[0086] ;

[0087] wherein, represents the standard value of the design standard template at the point with coordinates ; is the difference value; is the error value; is the weight value of the point with coordinates .

[0088] Specifically, by calculating the difference between the actual measurement point and the design standard point , which reflects the deviation between the actual state and the ideal state of the curtain wall at each coordinate position, and then combines with the weight function and the reflectivity compensation factor to further correct the error influence, so as to obtain the overall error . This method not only considers the local geometric deformation, but also weights the importance of different regions, making the error evaluation result more in line with the actual application requirements. During the actual construction process, due to various factors such as installation technology, material properties, and environmental lighting, the glass curtain wall often undergoes obvious deformation or offset in some local areas. Without quantifying these errors, it is difficult to determine whether the construction quality meets the design requirements. Through this formula, the deviation of each point can be accurately calculated, where is the actual measurement coordinate obtained after image compensation, feature extraction, and coordinate correction in the previous steps, while is the ideal coordinate preset according to the design standard template. Since in actual measurement, some points may be affected by optical distortion, environmental lighting changes, or reflection interference, resulting in noise or errors in their measurement values, it is necessary to calculate the difference of each point to identify the degree of actual deviation from the design. Next, the error calculation does not stop at the difference itself, but further reflects the true impact of the error by weighting the sum of squares of the differences.

[0089] The weight function plays a crucial role here. It is assigned according to the position, importance, and possible error sensitivity of each point in the curtain wall. Higher weights are often given to the edge areas or key joints of the curtain wall because these areas usually bear greater forces or are more obvious in visual effects, while the weights in the central or symmetric areas may be lower to balance the overall error distribution. In addition, in order to make up for the uneven influence caused by the surface reflection characteristics of the glass curtain wall on image acquisition, the reflectivity compensation factor is introduced into the formula. This is mainly to amplify the areas where the measurement signal is attenuated due to high reflectivity, so that the final error value can more objectively reflect the actual construction error. Specifically, the surface reflectivity and phase angle of the glass curtain wall will change under different incident angles . In the high-reflection area, the original signal may be overly suppressed. By introducing reflectivity compensation, the difference in these areas can be appropriately amplified during calculation to ensure that the error evaluation will not be distorted due to optical characteristics. Finally, by taking the square root of the weighted sum of squares of the differences, the overall error , this process actually combines all local deviations to form a global error index, which not only reflects the severity of local deformations but also the contribution of each region to the overall error. In the construction of glass curtain walls, this error index can be directly used to determine whether installation adjustments are needed or whether there are serious construction defects that require rework.

[0090] Example 9: Weight value It is represented by the following formula:

[0091] ;

[0092] Where, is the decay parameter of the spatial weight, which is a set value; is the sensitivity parameter of the feature intensity weight, which is a set value; ; is the central value in the design standard template.

[0093] Specifically, the weight calculation formula adopted by this method consists of two main parts, which respectively describe the importance of spatial position and the influence of feature gradients. The first term is a spatial weight term based on Gaussian decay: ; The physical meaning of this term is that according to the point and the center point of the design standard template, weighted in the form of a Gaussian distribution, so that points farther from the center have smaller weights, while points closer to the central region have larger weights. The parameter controls the decay rate of the weight. The larger its value, the lower the sensitivity of the system to spatial position. When takes a smaller value, the weight will decay rapidly, indicating that the system pays more attention to the central region. Since the construction errors of glass curtain walls may show certain distribution characteristics as a whole, for example, during the installation process, due to the influence of the support structure and fixed frame, the errors may be more obvious in local areas. Therefore, through this Gaussian weight term, the distribution of errors in space can be effectively reflected, and the error calculation can more accurately evaluate the overall deformation of the curtain wall. However, relying solely on spatial position for weighting is not enough because the errors of glass curtain walls not only depend on coordinate offsets but also are affected by surface feature changes. For example, at the edges or joints of the curtain wall, the connection method of materials may cause more obvious local deformations. Therefore, the influence of feature gradients needs to be considered additionally. For this purpose, the second weight calculation term based on feature gradients is introduced in this method: ; This term is used to measure the intensity of feature changes at the point and is weighted in an exponential decay manner. Here, represents the point The feature gradient at [location] can describe the local changes on the surface of the glass curtain wall, while the parameter controls the sensitivity of the feature weight. When takes a smaller value, it indicates that the system is more sensitive to local feature changes. When takes a larger value, the system's response to feature changes is weaker. Since the feature changes on the curtain wall surface often concentrate in the boundary areas or joints, the introduction of this weight item can effectively highlight the influence of these areas and reduce the error weight of flat areas, enabling the system to pay more attention to key parts during error calculation without being affected by uniform surfaces. By combining these two parts of weights, the finally calculated weight value can reflect both the positional relationship of points in space and the changes in their local features. Especially for the splicing area of the glass curtain wall, since the errors in this area have a greater impact on the stability of the overall structure, in this method, the weight of this area is often higher, thus ensuring that the error calculation can more accurately reflect the construction quality of the curtain wall. For the flat area in the center of the curtain wall, since its errors have a smaller impact on the overall structure, the weight value will be relatively lower, thereby reducing the influence of irrelevant areas during the error calculation process and making the system's calculation more efficient and accurate. The advantage of this weight calculation method is that it not only considers spatial factors but also combines the actual feature information on the surface of the glass curtain wall, making the error calculation more in line with the physical characteristics of the curtain wall. Compared with the traditional uniform weight distribution method, this method can more accurately identify the key areas of errors and improve the detection accuracy through a reasonable weighting method. In addition, since the weight calculation uses a Gaussian distribution and an exponential decay function, the calculation process has high stability and can adapt to different types of glass curtain wall materials and construction environments. In practical applications, this method can effectively improve the accuracy of error detection, enabling the system to not only remain stable in a high-reflection environment but also automatically adjust the weight distribution according to different curtain wall structures to ensure the rationality of the error assessment results.

[0094] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A glass curtain wall construction error detection method based on computer vision, characterized in that: The method comprises: Step 1: Obtain the original image of the glass curtain wall through a camera; use polarization imaging technology to obtain the surface reflection characteristics of the glass curtain wall; perform intensity reflection compensation on the original image based on the surface reflection characteristics to obtain a compensated image; Step 2: Extract feature vectors from the compensated image, and calculate the matching metric between each point in the compensated image based on the feature vectors; Step 3: Perform 2D reconstruction correction based on the matching metrics between each point and the camera calibration parameters to obtain the 2D point coordinate correction value corresponding to each point; Step 4: Compare the 2D point coordinate correction values ​​with the design standard template, analyze the deformation and evaluate the error.

2. The glass curtain wall construction error detection method based on computer vision as claimed in claim 1, characterized in that: The surface reflection characteristics of the glass curtain wall are obtained through the following formula: ； in, Indicates the incident angle of the glass curtain wall surface and phase angle Surface reflectivity under represents the refractive index of air; Indicates the refractive index of the glass curtain wall, ranging from 1.5 to 1.7; represents the incident angle of the light; represents the phase angle; Indicates the angle of the principal axis of polarized light.

3. The glass curtain wall construction error detection method based on computer vision as claimed in claim 2, characterized in that: The following formula is used to perform intensity reflection compensation on the original image based on the surface reflection characteristics to obtain a compensated image: ； in, Indicates that the compensated image is at any point The strength of the place; The original image at any point The strength of the place; is the reflectivity of the internal interface of the glass curtain wall; Ambient light at any point The strength of the place.

4. The glass curtain wall construction error detection method based on computer vision as claimed in claim 3, characterized in that: In step 2, the feature vector is extracted from the compensated image using the following formula: ； in, The scale is Gaussian filter; express The gradient of Represents any point The eigenvector at .

5. The glass curtain wall construction error detection method based on computer vision as claimed in claim 4, characterized in that: In step 2, the matching metric between each point in the compensated image is calculated based on the feature vector using the following formula: ； in, Indicate point The eigenvector at ; Indicate point The eigenvector at ; Indicates that the compensated image is at point The strength of the place; Indicates that the compensated image is at point The strength of the place; Represents L1 norm operation; Represents the sensitivity parameter of image intensity matching; if the glass curtain wall is high-transparency flat glass, The value range is 8 to 12; if the glass curtain wall is frosted glass, The value range is 12 to 18; if the glass curtain wall is coated glass, The value range is 15 to 20; if the glass curtain wall is colored glass, The value range is 18 to 25; Indicate point With point The matching metric of The average matching metric is obtained by calculating the mean of the matching metrics between all points .

6. The glass curtain wall construction error detection method based on computer vision as claimed in claim 5, characterized in that: Step 3: According to the matching metric between each point and the camera calibration parameters, two-dimensional reconstruction correction is performed to obtain the expression of the two-dimensional point coordinate correction value corresponding to each point: ； in, For any point The X-axis coordinate of For any point The Y-axis coordinate of Indicates the depth value; is the X-axis coordinate of the camera's principal point; Indicates the Y-axis coordinate of the camera's principal point; Indicates the position of the camera optical center; Indicates the location of the projector; Indicates the focal length of the camera; is the transpose symbol.

7. The glass curtain wall construction error detection method based on computer vision as claimed in claim 6, characterized in that: Depth value Calculated by the following formula: ； ； ； in, Represents the angle between the camera's line of sight and the optical axis; It represents the angle between the projector's line of sight and the projection axis; Indicates the baseline distance between the camera and the projector; Represents the wavelength of structured light.

8. The glass curtain wall construction error detection method based on computer vision as claimed in claim 7, characterized in that: In step 4, the following formula is used to compare the two-dimensional point coordinate correction value with the design standard template to analyze the deformation and evaluate the error: ； ； in, Indicates that the design standard template is at coordinates The standard value at the point of is the difference; is the error value; The coordinates are The weight value of the point.

9. The glass curtain wall construction error detection method based on computer vision as claimed in claim 8, characterized in that: Weight value Use the following formula to express it: ； in, is the attenuation parameter of the spatial weight, is the set value; Indicate point The characteristic gradient at is the sensitivity parameter of the feature intensity weight, and is the set value; ; is the center value in the design standard template.

Images